About the Authors John Lawson John Lawson is an e-commerce speaker and educator and the founder of ColderICE Media, a digital consulting company based in Atlanta, Georgia. John is the author of “Kick Ass Social Commerce for ePreneurs� due out in March of 2014; published by BenBella Books, Inc.
Contact Info: john@colderice.com Follow John on Twitter: @ColderICE
Sam Mallikarjunan Sam Mallikarjunan is the Head of eCommerce Marketing at HubSpot. Before coming to HubSpot, Sam worked as the CMO at lead generation and eCommerce firms where he focused on leveraging inbound marketing methods to improve marketing and sales efficiencies.
Contact Info: smallikarjunan@hubspot.com Follow Sam on Twitter: @Mallikarjunan
CONTENTS 1. Introduction 2. The Call-to-Action (CTA) 3. Dynamic Content 4. The Product Detail Page 5. The Top 10: Anatomy of a Product Page 6. How to Target Product Page Content Detail 7. Experimentation and Testing 8. Landing Page Conversion Rates 9. How to Design an Experiment 10. Conclusion
4
ECOMMERCE conversion optimization and testing
Introduction Sales in the eCommerce space continue to grow, and analysts believe the force of online retail is unstoppable. eMarketer research projects a 15% increase in consumer online spending by 2016, to an average of $1,738 annually. Seventy-one percent of consumers believe they can save money and receive a better deal by shopping online. For today’s connected and capable citizen, ecommerce offers an opportunity to maintain control over the research and purchase experience.
The growth of online retail has led to massively increased competition in the space. As a result, ecommerce companies heavily invest in expensive, purchase-phase outbound marketing that has caused the Cost of Customer Acquisition (COCA) to skyrocket while driving profit margins even lower. The brands who will win in the future are those who take a nimble approach to optimization, testing, and buyer persona profiling.
www.Hubspot.com
|
www.colderice.com
5
ECOMMERCE conversion optimization and testing
Developing narrative buyer personas of the people your company is trying to reach helps you shift the focus toward your prospects and customers. Gleanster Research has found that time engaged with relevant content is strongly correlated to likelihood of a sale. It’s not the quantity of interactions with a website that matter as much as the quality. Tailoring your landing pages and product pages to buyer personas is crucial to providing the relevant content modern consumers need to choose your brand above a competitor. In this ebook, you will learn the principles of effective calls-to-action that direct traffic to conversion points, the critical and testable elements of pretransactional landing pages, and the anatomy of effective product detail pages. You’ll also learn how to design and conduct experiments to improve your conversion rates. Finally, we’ll also address the growth of dynamic content, and how you can provide a personalized experience to repeat visitors and customers. There’s no shortage of data in the ecommerce space, but knowing how to leverage and apply intelligence for continual improvement is key to success.
www.Hubspot.com
|
www.colderice.com
6
ECOMMERCE conversion optimization and testing
The Call-to-Action (CTA) Today’s consumer is more distracted than ever before. Time Warner research has found that switching screens up to 23 times an hour is common, particularly among young people. Your website has to effectively compete against a number of other media options which fill your prospects’ browser tabs and other mobile-enabled devices. The need to fight to gain and maintain your website visitor’s focus has been referred to as an attention economy. Calls-to-actions are the tool to capture and direct attention and clicks to landing and product pages. Effective calls-to-action are one of the key differentiating web design developments of the last decade. Whereas early websites were like skyscrapers with a single front door (the homepage) and hundreds of floors that someone had to pass through to get where they’re going, modern inbound websites are like two-story buildings with thousands of front doors. A visitor can enter your site on any page through a search engine, social media referral, or referring inbound link, and the CTAs on each page should empower that visitor to find a relevant conversion point (transactional or pre-transactional) with a single click.
www.Hubspot.com
|
www.colderice.com
7
ECOMMERCE conversion optimization and testing
In many ways, call-to-action buttons function as a major tipping point on websites. When a website viewer encounters a CTA, they have the option of bouncing off your website or choosing to engage with your landing page. Optimization and testing are key to ensuring your CTAs sufficiently attract your visitors to keep them engaged with your website.
Primary vs Secondary CTAs While call-to-action buttons can promote a wide range of content offers and landing pages, there are two primary categories:
●. Primary CTAs: These buttons invite users to convert in a manner that is
directly tied to the sales process, which could include relevant product pages, information on how to reach a sales representative, or information on current discounts.
● .Secondary CTAs: Also known as pre-transactional conversions, these
CTAs allow you to capture the interest of website visitors who aren’t ready to make a purchase by promoting content offers, ebooks, or other tools which are helpful during the awareness and research stages of the sales cycle.
Both types of CTAs are critical to your contact and customer generation efforts, and using a mix of primary and secondary buttons will allow you to convert prospective customers, regardless of their stage in the sales cycle.
www.Hubspot.com
|
www.colderice.com
8
ECOMMERCE conversion optimization and testing
Calls-to-Action Click Through Rates Basic “button” calls-to-action on your site have five primary elements: the color, copy, shape, size, and value proposition. CTAs are typically thought of as a button, clickable elements on your site that take visitors to product detail pages or to nontransaction offer landing pages. These are certainly valuable pathways to test. However, ecommerce marketers often fail to test the most important Primary CTA on their website: the Add To Cart & Checkout buttons. Some HubSpot users have experienced up to a 1500% increase in the conversion rate of their product detail pages simply through A/B testing their Add To Cart button’s five primary variables. However, most marketers simply accept the Add To Cart button that came stock with their ecommerce website’s design without acknowledging that it’s a conversion web element for which they should take responsibility.
www.Hubspot.com
|
www.colderice.com
9
ECOMMERCE conversion optimization and testing
For both primary CTAs, like Add To Cart, and secondary CTAs on your landing pages, you should again focus on the most disruptive elements first. The primary color palette is a good starting place for testing. In general, you want your CTAs to violate the color scheme that surrounds them. They should be visually distracting and an obvious target for the desired point of engagement for visitors. However, exactly which color is most effective is going to be unique to each website and is something that can only be determined by testing and experimentation. Be warned that professional graphic designers could find themselves suffering great emotional distress or conflict during this phase of testing, since a good CTA, by its nature, is going to be visually disruptive in a way that may be uncomfortable for the aesthetically-inclined. After working your way through variations on the primary color palette, you can work on the copy of the CTA itself. The copy isn’t necessarily what you’re saying, it’s how you say it. Punctuation, sentence structure and diction can influence how effectively you’re communicating the value proposition of why someone should click. When testing this feature, it’s important to also track the click-to-conversion rate to ensure that you’re not abusing the copy to increase the click-through rate (which is a good and valuable metric) at the expense of the landing page’s conversion rate.
www.Hubspot.com
|
www.colderice.com
10
ECOMMERCE conversion optimization and testing
At the end of the day, the conversion rate is the far more important metric. A high click-through rate and low conversion rate can be indicative of a call-to-action that has set inaccurate expectations for your content offer. For example, you could relabel your Add-to-Cart button to say “Click here to get a free puppy!” and probably massively increase your click-through rate (who doesn’t like puppies?) but the conversion rate of the shopping cart would plummet and the user experience would be severely disrupted.
The shape of the CTA should also be disruptive from the overall design of your page. In general, most websites use square or rectangular CTAs with either edged or rounded corners, but that doesn’t mean that’s the most effective shape for your website. You can test any number of shape configurations to see what draws the attention of the visitor’s eye, while still communicating the fact that it’s a call-to-action. Users of the internet are programmed to innately understand that standard button shapes are clickable, and while designing your CTA in the form of a Tyrannosaurus Rex may sound like a cool idea, it could actually lower the clickthrough rate by preventing people from recognizing it as a CTA. However, we’re not going to say this is absolutely true, and if you want to test a T-Rex iteration as part of your due diligence as a marketer, then that is your prerogative. www.Hubspot.com
|
www.colderice.com
11
ECOMMERCE conversion optimization and testing
The size, similar to the shape, is a relatively standard function of most website’s designs. In general, clickable web elements that are larger are more disruptive of the graphical layout of the page and will get more clicks. However, in some cases smaller CTAs that add more white space around them might get a higher CTR. This is another variable whose impact will change based on your unique site design and the layout of your web elements, and is worthy of testing. Value proposition, as opposed to copy, isn’t just how you’re saying something but what you’re actually saying the value of the offer on the landing page is. Again, this is a variable which you should monitor to ensure that the click-to-conversion rate isn’t negatively affected. For example, you can alter the CTA to say “Click Here for Your FREE Pet T-Rex!” and probably get a higher click-through rate based on the high perceived value of that offer, but when it takes them to a landing page with copy specifying that they’re downloading an ebook about your product’s industry, the conversion rate of the landing page will suffer. However, you can here test the rhetorical appeal of the value proposition as well as what you emphasize as the valuable part of the offer. www.Hubspot.com
|
www.colderice.com
12
ECOMMERCE conversion optimization and testing
Of course, in addition to standard-design button-type CTAs, there is an infinite universe of combinations for CTAs with buttons inside buttons and varying designs. The five basic variables we’ve discussed here are fairly universally applicable, but your particular design may have many other disruptive variables that you can - and should - test.
www.Hubspot.com
|
www.colderice.com
13
ECOMMERCE conversion optimization and testing
Dynamic Content Dynamic content, also known as smart or adaptive content, is when the content of a website or email adapts to a user’s behavior and preferences. This process is achieved using insights gleaned from their interactions with the brand’s website to date. Amazon offers one of the best-known dynamic content experiences. Their recommendation engine provides personalized recommendations to repeat shoppers, based on a number of factors which can include purchase history and onsite product page engagement. For online retailers, offering dynamic content is key to emulating the personalized experience shoppers crave. Jupiter Research has discovered that relevant email content can drive 18 times more revenue than messages sent to a non-segmented list. Dynamic content is only possible with the help of technology, and by obtaining the tools to compile, sort, and apply sufficient insights to get the content and the timing right. Instead of your website being a “one-size-fits-all” soapbox, it should become a dynamically updating user experience that conforms itself to the customer’s buyer persona changing the offerings and CTAs they see based on their pain points, past history, and other engagement and contact data.
www.Hubspot.com
|
www.colderice.com
14
ECOMMERCE conversion optimization and testing
Practical applications of dynamic content The primary purpose of dynamic content should be to improve your prospect experience. When used indiscriminately, smart content has the potential to confuse or even disturb website visitors. Collecting behavioral and historical insights is only the beginning of how this practice should be applied:
●. Remove Repeat Conversions: By compiling insights on each website
visitor’s conversions to date and tying these insights to your CTAs, you can
eliminate the chances they will be shown an offer they’ve already converted on. Smart CTAs increase your chances of repeat conversions, and create a fresh and exciting experience for your prospects.
● .Map Offers to Their Lifecycle Stage: With the help of a centralized
database which tracks your prospects’ and customers’ stages in the buying cycle, your smart content can be tailored to their buyer readiness. Leads who have only converted once or twice can engage with content offers
tailored to the research stage, while loyal customers can be shown exclusive promotions. You can also collect simple qualifying data (such as what a
prospect would use a specific product for or when they’re planning to make
a purchase) and customize the content they see on the website to appeal to those specific qualifying characteristics.
www.Hubspot.com
|
www.colderice.com
15
ECOMMERCE conversion optimization and testing
â—? .Leverage Progressive Profiling: Dynamic content can immensely benefit
ecommerce marketers who utilize smart landing page forms for progressive profiling, the practice of gradually compiling additional insights on your
contacts. Depending on your insights on the customer, they can be shown
a shorter form, which asks new questions so you can enhance their profile, or bypass the form entirely. This is particularly useful for nurturing existing customers with content offers. You can use a generalized content offer
with a few dropdown options after a sale that asks questions around how
a customer plans to use a specific product, and based on that, customize their experience (including selling related products and accessories) to an
increasingly detailed picture of who that customer is as a person and what needs you’re helping them address.
www.Hubspot.com
|
www.colderice.com
16
ECOMMERCE conversion optimization and testing
The Product Detail Page As an online retailer, the product page is where you must close the sale. It is simply the most important part of the entire web site. However, it may also be the most overlooked. Many ecommerce companies rely on cookie-cutter product titles and dry, technical descriptions lifted straight from the manufacturer. Even more common is a total reliance on the default page layouts set up by the web design team. Remember, if you’re just starting out in inbound marketing and only attracting purchase-ready visitors, a web user may never see any other part of your site except for the product page. They may never venture to your home page, a landing page, or any other page on your site. A keyword search of a product’s name or attributes on Google will often direct consumers straight to your product pages. As search engines become more adept at searching for relevant content, organic search traffic for purchase-phase keywords is more often directed to product pages. Considering the user experience of a first-time website visitor who has landed on a product page is crucial. If the page is not designed to convert and optimized to close the sale, you could be leaving a lot of money on the table.
www.Hubspot.com
|
www.colderice.com
17
ECOMMERCE conversion optimization and testing
The typical conversion rates for an internet retail site are anywhere from
1.5% - 3.5% If you could close 50% more of your product page visitors, your top and bottom lines would increase dramatically. An item page that’s optimized to close has the potential to double your income more quickly than any other form of optimization you may decide to do. Retailers may spend tens of thousands of dollars optimizing their websites for SEO in order to receive more traffic, but not spend nearly enough time and effort on testing their product pages.
www.Hubspot.com
|
www.colderice.com
18
ECOMMERCE conversion optimization and testing
The Top 10: Anatomy of a Product Page
1. .
Search Box / Search Bar
Analyze your visitors’ search terms regularly, and keep close tabs on the the keywords that drive traffic from search engines to your website. Why? Customize the keywords in your product descriptions to closely match the terms that your customers are using to find your product pages through search engines and on your website.
www.Hubspot.com
|
www.colderice.com
19
ECOMMERCE conversion optimization and testing
You might end up wasting a lot of time by using keywords that people are not using to search for your products in the first place. Applying insights from your search bar and website metrics requires making changes to your product page content in response to search trends. You should examine which keywords are driving the most valuable traffic to your site as well. “Closed-loop” marketing is the practice of analyzing traffic sources and seeing what sources convert at the highest rate. If you’re receiving a lot of traffic from a specific keyword but no conversions, you may want to optimize for a lowervolume but higher-converting keyword.
www.Hubspot.com
|
www.colderice.com
20
ECOMMERCE conversion optimization and testing
2. .
Product Titles
Treat your product titles like the lock to Pandora’s box: concise and compelling descriptions are the key that welcomes visitors to the page and entices them to continue reading the rest of your content. ●. Titles are giveaways to what your page contains. Writing a title that represents
the main descriptor behind your product is important).
● .Titles capture your visitors’ attention. Devising a title that that is not too
lengthy, has an element of catchiness, and contains enough information to adequately describe the scope of your product is important. ●. Titles equate to relevance. This means titles should be very relevant to
the product on page, as well as why your buyer personas are considering purchasing the product.
●.Titles represent your business. The manner in which your title is written says
much about your business. Titles that contain inappropriate language, are incorrectly spelled, are erroneously punctuated, or poorly designed leave an undesirable impression
Best Practice Tip Search engines allow up to around 65 characters of titles, but longer is not necessarily better. Strive to keep your titles to around 40-50 characters. The more characters in your title, the higher the probability that search engines will truncate it when displayed in the search results page.
www.Hubspot.com
|
www.colderice.com
21
ECOMMERCE conversion optimization and testing
3. .
Product Page Images Links are important to the search optimization of text-based content pages. Much like text, images can also be optimized to appear more often through organic searches.
Images can be found through the keyword you use and through an organic
search, and they may show up through image search or in the universal results page. Note that your goal in using images on your site should not only be to
drive traffic, but also boost sales and increase the chances of sharing through visual-content-heavy social networking sites like Facebook, Google+, and
Pinterest. Much like for web pages in general, the image URL, alternative text, and tag descriptions are also essential for optimizing images.
Keep images on the product page very relevant to the page content and
purpose of the product; an image of an eye could be used for an eye-related item. Additionally, save and rename your files to include a keyword that’s
relevant to your product before adding to your product pages. The filename for that image should contain a keyword related to the same product.
Best Practice Tip Use images with high quality. Clear and sharp images are important in order to not distract your visitor’s focus. User experience research has found that quality images are critical to first impressions of your web page, while text compels visitors to continue engaging with the page.
www.Hubspot.com
|
www.colderice.com
22
ECOMMERCE conversion optimization and testing
4. .
Rating / Reviews Much like in the realm of traditional customer service, you can’t afford to ignore your customer’s voice when describing your product. You’ve surely noted how social media network users often post about trending topics, and content with viral qualities earns the most ReTweets and shares. The same principle should be applied to user-generated content on your product pages.
Unbiased feedback from other consumers who’ve purchased the product acts as social proof, which first-time website visitors use to gauge the legitimacy of
your brand and products. Many companies edit out negative reviews and leave only positive reviews, which is a mistake. The occasional negative review is
critical to giving the positive reviews credibility. Consumers are smart, and they know that it’s implausible that every single customer has always been thrilled. They’re willing to overlook occasional negative reviews and lend a higher
credence to the positive reviews if they think they’re authentic and organic. If your site is receiving such a large volume of negative reviews that they’re
crowding out positive reviews, it may be indicative of a need to better identify and empower promoters to leave positive reviews, or it may be indicative of service-related issues within the business.
Remember, user-generated content is not only about text. You should welcome the sharing of pictures, videos, and even audio on your site, if applicable. The more opportunities you can accommodate, the more users will get involved.
The more user-generated content you are able to incorporate, the more your
conversion rates will also increase accordingly. Personification is an incredibly
powerful tool, and if you can empower user-generated video or audio, they may have an even greater impact than simple text or image reviews.
www.Hubspot.com
|
www.colderice.com
23
ECOMMERCE conversion optimization and testing
5.
Price Inbound marketing experts have yet to reach a consensus about whether
pricing information is beneficial to the websites of offline retailers or other
businesses. There’s compelling evidence in both directions, and data to support both positions. Many businesses are opting against displaying pricing in order to leave room for negotiation or customized packages.
However, in the retail ecommerce space, there should be no confusion over
pricing best practices. If you don’t place accurate and clear price information
on your product pages, you stand a chance of not setting proper expectations among your customers. Retail consumers don’t like surprises, and your
bounce rates will soar if they’re not able to efficiently find answers to their top questions. One of the leading causes of shopping cart abandonment is the
so-called “sticker shock” of seeing the combined price of multiple orders plus
shipping costs. Minimize the shock by setting clear expectations throughout the browsing process to reduce the emotional stress of the checkout experience.
If your rates are different than the cost of the same item in a brick-and-mortar retail space, detail the disparity to communicate value to your page visitors.
Best Practice Tip Items that are on sale or discounted should show both the original price and the discounted amount to create a sense of urgency and enhanced value in prospective buyers.
www.Hubspot.com
|
www.colderice.com
24
ECOMMERCE conversion optimization and testing
6. .
Shipping Info Customers do not like a big surprise when they view shipping costs at
checkout. You should manage expectations by clarifying shipping and handling
costs up front. Some websites offer clients real-time calculation tools, based on a small number of fields. This can mitigate some cart abandonment, and allow you to present a transparent front to your product page visitors.
7. .
Add to Cart Button The Add to Cart is the primary conversion CTA on most retail sales pages. If you can get a prospect to place an item in their cart, your product page has effectively converted a page prospect. Execution of the purchase depends
on the user-experience of your checkout process, which should be analyzed
independently of the product detail page so that you can focus on optimizing the product page to convert prospects into the checkout process..
On your product pages, be sure that your cart button is prominent. Generally,
the cart button should be distinct, and in a color that stands out over any other
buttons on your page. Place it beside each product, and move it above the fold so it’s not difficult to find. There is room for other call-to-action buttons on your pages, which can allow visitors to complete secondary conversions or invite
them to socially share the page links in order to generate more traffic. However, the placement and color of social sharing buttons or secondary call-to-actions shouldn’t compete with the Add to Cart option.
www.Hubspot.com
|
www.colderice.com
25
ECOMMERCE conversion optimization and testing
8. .
Product Description Some online entrepreneurs settle with using pure images on product pages,
mistakenly assuming that no one will read their product descriptions. Others rely on the manufacturer’s product descriptions, without making any modifications. In both instances, your chances of appearing in search results are negatively impacted by not having an effective, relevant product description that adds to the context of the page. Search engines use content to help identify the
best page to present a searcher based on the context of the copy, and by only having a product title and failing to invest in unique, comprehensive product
descriptions you’ll significantly reduce your chances of standing out from your competitors.
Avoid the temptation to use lengthy product descriptions in order to incorporate the largest amount of keywords possible. Prioritize primary keywords, and don’t force the use of secondary keywords if it extends the text or results in unnatural copy. Make your content as relevant to your buyer personas as possible by
using simple terminology, and most importantly, writing in language that appeals to your buyer personas.
www.Hubspot.com
|
www.colderice.com
26
ECOMMERCE conversion optimization and testing
Always focus on a natural-sounding product description that focuses on
delivering a valuable experience to the prospect. There’s a whole small city’s
worth of smart folks in California focusing on identifying and severely punishing marketers who focus on optimizing for search engines at the expense of optimizing for the user.
Writing 300 or more product descriptions may be tedious, but it’s critical to
avoid content that’s redundant in order to maximize your chances of ranking and prevent being actively penalized by search engines for using duplicate
content. Introduce variation in order to optimize for the keywords which are driving the highest volume of search in your niche.
9. .
Alternative Images Multiple images and views of your product will allow consumers a more
holistic experience when deciding whether or not to purchase that product.
Different angles, different configurations, and different product uses are key to empowering users to see the features that are important to them. One of the
key remaining hesitations of consumers in adopting ecommerce has been their
inability to have the experience of holding a product and seeing how it interacts
with its environment like they can do in a physical retail store. By adding photos that show a product in more than the simple, clean light of the main product
image, you can help your prospects literally get a clearer picture of the product and feel more comfortable about a decision to purchase.
www.Hubspot.com
|
www.colderice.com
27
ECOMMERCE conversion optimization and testing
10. .
Social Sharing Buttons An optimized web page will include icons that link to sharing on different social media networking websites. Provide visual links to every network, including
Google+, Facebook, Twitter, Reddit, LinkedIn, Pinterest, and more. Use your
buyer personas’ social media networking habits as a guide in determining which networks to include. Incorporating these icons will provide a convenient option
for social sharing by your page visitors, maximizing your chances of exposure in their networks.
Empowering visitors to share your product detail pages is critical for several reasons. The most
basic reason is that social shares
allow you to gain penetration into the social networks of your site
visitors. In addition to the traffic,
social referrals may have a higher
chance of turning into customers. Social recommendations can be an extremely powerful influence
on a consumer’s decision to buy. Another important reason to empower and encourage social sharing is the
increasingly massive level of importance that search engines are placing upon social shares. It makes sense, of course. With more than 1/8th of the entire
world’s population manually going through the internet saying “I think this is
interesting and authoritative enough to share with my friends,” it was inevitable that search engines would start to heavily weight social signals. It’s difficult for a purely on-page oriented algorithm to make the quality of results driven by analyzing social engagement.
www.Hubspot.com
|
www.colderice.com
28
ECOMMERCE conversion optimization and testing
Also, social shares have heavily democratized Search Engine Results Pages (SERPs) rankings. When inbound links were the primary metric of SERPs, only webmasters or people that maintained their own websites on a free
CMS like Blogger could “vote� on what made an authoritative site or page by
linking to it. Since social accounts are far more prevalent than people actively maintaining websites and controlling outbound links, a far greater number of
people can now contribute to the equation of what makes a given website more authoritative and quality than another.
Best Practice Tip In the eCommerce realm, product photos are a critical component of the item page. Consider placing your social sharing buttons near the photo, to allow website visitors to conveniently share the image on a social network.
www.Hubspot.com
|
www.colderice.com
29
ECOMMERCE conversion optimization and testing
How to Target Product Page Content Detail As your prospects and repeat customers navigate your website, they view your product pages with a very specific question in mind: “What’s in it for me?” More technically, your website visitors engage with product pages to evaluate how well the item will solve their unique challenges and needs. As an eCommerce retailer, it’s optimal to create multiple product landing page iterations for each product, which are tailored to your buyer personas’ unique needs using dynamic content. Gleanster Research has found that a majority of B2B brands are currently investing in the tools necessary to integrate dynamic content.
Personalized product pages may become the norm in the near future. An optimal online retail experience will integrate product pages into a dynamic content system that has at its center the history of a specific customer contact record to address the individual needs of your customers. Create multiple product page iterations using dynamic content to provide buyer personas with the most relevant experience possible.
www.Hubspot.com
|
www.colderice.com
30
ECOMMERCE conversion optimization and testing
The following factors can aid in tailoring your product pages:
Address Problems Directly Map your buyer personas’ problems to your product, and explain how the
product can act as a solution to their needs. This is particularly powerful if
you’ve gathered qualifying information during a pre-transactional secondary conversion around the pain points and use cases the prospect is looking to
solve. If they’re planning to use their HDTV primarily to watch sports, you might emphasize different key features than if they had indicated that they were planning to us the HDTV for gaming. Play Up Priorities Are your buyer personas primarily concerned with their budget, the quality of
product, or convenience? Include copy and components which detail how the purchase would meet their specific needs. Prior Purchase Is the product an add-on, enhancement, or natural addition to something the
customer has already purchased? Clearly demonstrate the relationship between the two products. User Experience Adjust the web design, content, and user experience of your product pages to your buyer personas’ level of savviness, which can include enhancing or
removing technical jargon or making the web page easier to use by simplifying options versus having robust customization options.
www.Hubspot.com
|
www.colderice.com
31
ECOMMERCE conversion optimization and testing
Comprehensive buyer persona profiles will include sufficient insight on demographics and psychographic characteristics to allow you to step into the position of your prospects during the product research and purchase process. Evaluate your product pages with their unique needs and priorities in mind in order to create relevant dynamic content.
www.Hubspot.com
|
www.colderice.com
32
ECOMMERCE conversion optimization and testing
Experimentation and Testing The only intelligent way for marketers to improve the performance of their funnel metrics is to conduct tests and experiments. Since the dawn of professional marketing, we’ve relied primarily on opinions and gut-feelings to make decisions about how to improve the performance of our campaigns. Simply put, the technologies and methodologies to effectively conduct marketing experiments didn’t exist in the pre-internet world. It wasn’t feasible to conduct tests to increase the performance of billboard or television ads, for example. The media providers didn’t have the infrastructure built to serve different samples of a TV commercial to randomized segments of an audience, and the response to these ads was typically so low and so difficult to attribute that there wasn’t a significant, measurable metric by which we could have analyzed the impact anyway. Unfortunately, much of this same psychology has trickled into the modern era where marketers are too inclined to “go with their gut” rather than conduct proper, methodical experiments with significant metrics to improve their performance. This is one of the greatest dangers in modern marketing organizations, when marketers ignore data entirely or simply accept the performance of their efforts as they best they’re going to get without trying to improve. Simply put, it’s a lazy and behavioral relic of the Mad Men era when modern marketers ignore the fact that ecommerce websites have the ability to track almost every possible behavior and marketing activity that influences and drives them. www.Hubspot.com
|
www.colderice.com
33
ECOMMERCE conversion optimization and testing
Although there are a number of highly skilled and qualified researchers conducting ongoing experiments and analysis to develop “best practices” for inbound marketing, there are significant differences in the sample populations of each individual website that make the blind application of “best practices” ineffective. Since all of these experiments rely on behavioral analytics, and every website’s buyer personas respond, by their nature, in different ways to messaging and stimuli, it’s impossible to say that what works for one website (or what works when massive sets of data from many websites are analyzed as a whole) will work specifically for your website. Therefore, it’s more important that you use the data and methodology from these researchers to design and conduct experiments on your own site rather than simply designing your interfaces and marketing efforts based on the results of experiments conducted on unrelated consumer sample populations.
www.Hubspot.com
|
www.colderice.com
34
ECOMMERCE conversion optimization and testing
WHAT IS VARIATE TESTING? Multivariate testing is a controlled method of testing complex systems. In the marketing realm, the process allows you to test more than one section of your website in a live environment. Most often, A/B tests are used to determine which of two varying website components elicits the best reaction from website visitors over time. By using a platform which allows you to air two versions of content or creative components to an equal proportion of website visitors, you can gain insights on which variable wins more clicks, subscribes, or purchases. The only limit to the number of variables you can test is the amount of time it will take to receive a statistically significant sample of each variable. When using a dynamic content management system, you can collect information on user behavior following exposure to each variate you are testing.
www.Hubspot.com
|
www.colderice.com
35
ECOMMERCE conversion optimization and testing
WHAT IS A/B TESTING? The most fundamental experiment is the single-variate A/B test where one version of the element or tactic (called the “control” or “Version A,” which is often the existing version) is served to a randomized subset of the overall population while a second version (called the “treatment” or “Version B,” which has a single element modified from the control) is served to another subset of the same population. There are two very important factors when conducting variate testing in order to keep the experiment “clean” and have the results be significant and accurate.
First, you should ensure that each version or treatment you’re testing has only one element that has been modified from the control. If you change multiple elements - for example, the headline of a landing page, AND the number of forms, AND the image on the page, AND the structural layout of the page - you won’t know which of the elements that you modified was the element that drove the change in behavior. You can test multiple variables, but only if you do so methodically by creating iterations of the multiple combinations of variables. This method will allow you determine if there’s a specific mix of variable changes that are driving the greatest impact on the outcome of the experiment www.Hubspot.com
|
www.colderice.com
36
ECOMMERCE conversion optimization and testing
In order to drive the highest velocity of change and be able to reach levels of statistical significance as rapidly as possible, you should focus on testing the most disruptive elements first and making decisions as soon as you reach an acceptable level of statistical significance. For example, it makes sense to test a red vs. blue color for a call-to-action, which is likely to drive a significant change in behavior, before testing whether an exclamation mark drives more clicks than having no punctuation (which, if it drives any change in behavior at all, will be so small that the experiment will have to have a massive sample size before drawing a conclusion). At the other end of the spectrum, it’s not always necessary to let an experiment run for an excessive period of time and refine the experiment data to an extremely precise confidence interval before ending the experiment, drawing conclusions, and testing another variable.
Every test that you conduct should have a hypothesis and a key metric. For example:
“Changing The Landing Page’s Image Will Improve The Conversion Rate By At Least 5%”
www.Hubspot.com
|
www.colderice.com
37
ECOMMERCE conversion optimization and testing
Landing Page Conversion Rates The key performance indicator for pre-transactional landing pages, such as download pages for ebooks or Buyer’s Guides, is the conversion rate percentage. This metric is a measurement of the number of times the form on the landing page is filled out and submitted, divided by the total number of visitors to the landing page. For example, if a landing page gets 10,000 visits and 1,000 visitors fill out and submit the lead generation form, the conversion rate of the landing page is 1,000/10,000 or 0.1 or 10%. The assumption is that if all variables remain the same, one out of every ten people who visit that landing page will fill out the form and become a prospect lead that you can nurture. Improving this metric is a function of the design and messaging of the page, which has some specific variables that you can test. Consider the following: Layout As on product detail pages, one of the most disruptive elements of a pre-
transactional landing page is the layout of the various elements. The position on the page of the headline, the form, the image, testimonials, the copy, and other major components can all influence how quickly a visitor processes
the information and makes a decision about conversion. The configuration of
elements horizontally and vertically can also make a difference. Should the page copy be on the left or below the image so it is read after the image is seen but before the form is seen? Should the form be above the “fold� (the point where a visitor has to scroll to see it) so that the conversion purpose is obvious? By testing the layout and configuration of the various elements of your landing
page, you can change the order in which a visitor processes the information and influence their decision-making process in converting. www.Hubspot.com
|
www.colderice.com
38
ECOMMERCE conversion optimization and testing
Product Image We’ve previously talked about the impact of imagery and visual content on
rapidly communicating messages and value propositions to website visitors. First-time visitors to landing pages undergo a very brief period of evaluation
before deciding to convert, and the “hero image” (the main, disruptive image
on the page) is a key element of rapidly communicating your persuasive value proposition. Should the image be of a graph of product performance? Should
the image be of a person that’s reflective of the visitor’s buyer persona? Should the image be bright and loud, or dark and subdued? Should the image have multiple elements, such as star-outlined copy emphasizing the value of the offer? Many variables unique to your specific landing page can be tested
with the hero image. Due to its ability to communicate a persuasive value
proposition so rapidly, it should be among the first of the disruptive elements that you test.
Headline / Product Title The headline, along with the hero image, is one of the most important visual
elements of the landing page. It has obvious influence over a visitor’s decision process about the purpose of the landing page and its value proposition.
Should the headline be a statement or a question? Should the headline be the title of the offer or a short sentence about the consumer pain point on which
your content offer provides information? By testing the headline of the landing page to determine the most persuasive and descriptive text possible, you can
significantly influence how quickly a visitor comprehends the value proposition and improve the percentage of visitors that convert. www.Hubspot.com
|
www.colderice.com
39
ECOMMERCE conversion optimization and testing
Call to Action Button The form’s submission button can also influence the conversion rate of the landing page. It can be used to more clearly define the result of the action
(“Download Your Free Guide Now!”). Alternatively, it can appeal to the problem the offer addresses (“Learn How To Plan The Perfect Party”). Finally, the
submission button text can be used to ask a question (“Are You Ready to
Party?”). The standard copy and design is of a simple button, which is usually grey, that just says “Submit.” However, many website owners that test this
page attribute find that landing page visitors don’t like to “Submit.” In addition, a small button with a subdued color such as gray might not be attractive
enough to draw the visitor’s eye and incentivize conversion. As with all of these disruptive landing page elements, it’s possible that your optimized conversion structure will have the default button configuration. However, many online
retailers find that editing the messaging and design of the submission button can positively impact conversion rates. Copy Organization Since visitors to a landing page have a limited amount of time to consume the persuasive information about what the offer is, you should test the structure
and organization of the descriptive copy on the page. Should the copy be short and sweet or long and detailed? Should the copy be bulleted, enumerated, or in paragraph form? Should the paragraphs be broken up in different ways?
Copy on the landing page is one of the deeper, more influential elements to
those visitors who aren’t going to be influenced by optimization of the headline, image, and form. Your copy is among the best opportunities you have to
make a persuasive case to your prospect. By testing the length, type, and
configuration of the copy, you eventually develop an accurate barometer of
how quickly your visitors can process information before making their decision about converting
www.Hubspot.com
|
www.colderice.com
40
ECOMMERCE conversion optimization and testing
Value Proposition The value proposition of the landing page is directly related to how you position the reason. Have you effectively convinced your prospects to exchange
their contact information for your special offer? While most of the individual
elements of the landing page can be tested to determine how well they support the value proposition on a micro level, on a macro level the entire landing page has a value proposition that you should actively spin, emphasize, and refine. Ecommerce companies often excel at crafting brilliant value propositions for
product detail pages because they’re creating content with the understanding
Si
on
S
PoS
Se
S
that they’re selling something to the page visitor.
in f o rM
aT
io
n
Marketers should think of their landing pages like product detail pages, but with the understanding they’re selling their content offer. You’re offering an
item of value in exchange for something of value to the consumer (their contact information). The more and more valuable the information you’re asking for, the higher the price point of the investment you’re asking a prospect to make. A
form that requires age or phone number should only be attached to a page with a very clearly defined, enticing value proposition.
However, on the flip side, the greater and more convincing your value
proposition, the more you’re offering in exchange for their contact information. You can, and should, A/B test your value propositions to see which method of framing has the most positive impact on your conversions.
www.Hubspot.com
|
www.colderice.com
41
ECOMMERCE conversion optimization and testing
Rhetorical Appeal There are three primary modes or appeals of rhetoric: Logos (the logical
appeal), Pathos (the emotional appeal), and Ethos (the credibility or authority appeal). The rhetorical appeal which proves most persuasive for your site visitors is very much a function of the psychographic dimensions of your buyer personas. Should you emphasize on the landing page a detailed,
logical argument for converting on the page? Should you make an emotional proposition for how the offer will help solve their problems and challenges?
Should you emphasize the credibility of the data the offer is based on, or the credentials of its author?
A/B testing which rhetorical appeal has the greatest impact on conversion rate,
and then correlating the results to the form data that you’re using to identify the buyer persona serves several purposes. Not only can you better optimize the
way you structure your landing pages, you can develop a richer understanding of your buyer persona’s psychographic characteristics. By identifying their
responsiveness to each form of rhetorical appeal, you can apply this principle in other areas of your marketing to that buyer persona.
www.Hubspot.com
|
www.colderice.com
42
ECOMMERCE conversion optimization and testing
Social Proof Social proof and testimonials or reviews of the content can have a significant, positive impact on the conversion rates of landing pages, depending on your buyer persona. There are exceptions to this, such as early adopters, which
are a prime example of buyer personas which may not be easily swayed by social proof. You should test the effectiveness of different types of social
proof on the overall conversion rate of the page. Should you group social
proofs together? Should you have just one? Should you have none? Should you use reviewers who reflect a particular demographic profile? Testing all of these questions and correlating the results to the identified personas
based on form field submissions is another method of refining psychographic persona characteristics.
www.Hubspot.com
|
www.colderice.com
43
ECOMMERCE conversion optimization and testing
How to Design an Experiment It’s been said that “if you’re not measuring, you’re not marketing.” Taking an agile approach to designing and launching experiments is a key differentiator in the ecommerce realm. Before you begin tests to optimize your landing pages, having a defined goal in mind is critical. Are you aiming for more email subscribers, higher conversion rates, or more purchases? Without clear parameters, it’s difficult to collect and apply data.
www.Hubspot.com
|
www.colderice.com
44
ECOMMERCE conversion optimization and testing
1. .
Baseline VS Split Testing
Baseline testing is the process of changing a single element of your website to determine how it affects your conversion rates. By switching the background color of a call-to-action button after one month of using it on your website homepage, a baseline test will allow you to distinguish how your click-through rates improve over the previous iteration. While these types of experiments are certainly better than stagnancy, the accuracy of the insights is difficult to determine due to the fact that your website visitors and external factors, like time of year, can significantly change between the two tests.
december
In A/B or split testing, you’re able to determine the relative efficacy of two iterations by testing both simultaneously. Instead of having to wait to collect a statistically significant sample size before testing changes, the website variations can be tested in a live environment. The benefit to a well-designed A/B test is that you’re able to collect data by testing a variation of a current element, allowing you to begin optimization even if you haven’t been tracking metrics closely to date.
www.Hubspot.com
|
www.colderice.com
45
ECOMMERCE conversion optimization and testing
2. .
Calculating Necessary Sample Size
Even if the variation you’re testing results in an immediate spike in favorable behavior among your website visitors, it’s critical to continue your experiment until a statistically significant sample size is obtained. Determining sample size is an important first step towards any form of experimentation. A sample that is too small will not allow you to calculate actionable intelligence, while too large of a sample can waste resources and delay you reaping the benefits of deploying the optimized iteration and conducting experiments on other page elements. Sample size can be calculated in excel or by hand, using the following formula:
Aim for a confidence level of 95%, which means you can be 95% certain that your margin of error is accurate. To calculate sample size in excel with a 95% confidence level, use your average monthly website visitors as the population: ●. Enter your website visitors in cell A1. ●. In Cell A2, enter the following formula: A1/(1+(A1 * 0.05^2))
If you were averaging 2500 visitors per month, you would need a sample size of 345 in order to achieve actionable intelligence with 95% certainty that your margin of error is just 5%. Alternatively, you can use the free HubSpot Super Tool. www.Hubspot.com
|
www.colderice.com
46
ECOMMERCE conversion optimization and testing
Conclusion Online retail has changed the purchase process as we know it. While sales have traditionally been guided by human sales reps who are able to efficiently address objections and priorities, your online presence is now responsible for providing the information your prospects need at every stage of the buying cycle. The first step toward designing a web presence that’s optimized to the customers your company wants to acquire is developing specific narrative buyer personas, which can reveal sufficient insight to create relevant content that will attract and influence them. Optimization can be a key differentiating factor for online retailers; eConsultancy has found that for every $92 spent acquiring customers, only $1 is spent converting them. Success in ecommerce isn’t about collecting the most data, it’s about analyzing the right metrics and quickly applying insights. A/B testing, on average, results in 2025% higher conversion rates for ecommerce retailers. Regardless of your baseline traffic at the time you begin optimization, you can achieve significant results by calculating the appropriate sample size. eConsultancy has found that success rates are incredibly high among marketers engaging in testing; 99% of all those using experiments report satisfaction with their results.
www.Hubspot.com
|
www.colderice.com
47
ECOMMERCE conversion optimization and testing
It’s crucial to perform continual optimization experiments on your landing pages and product pages to apply elements which can significantly increase the success of your web pages. The best marketers continually test single elements in order of disruptiveness, because total optimization simply isn’t possible unless you have a completely controlled population. Continue testing until you’ve received a statistically significant sample size and low margin of error, and launch another experiment. The marketers who build lasting relationships with prospects and customers are those who leverage dynamic content technology to design a user experience that’s guided by relevant content. Tailor your experience to people with the help of a centralized marketing database which allows you to make personalized recommendations based on the holistic history of a specific contact’s interactions with your brand. The customer experience in the ecommerce realm should be personalized, and the companies who win will be those who can create a scalable system for presenting a highly relevant and enjoyable user experience.
www.Hubspot.com
|
www.colderice.com